Carbohydrate structures on proteins and lipids play critical roles in mediating and modulating biologically important protein and cell interactions. The precise localization of glycosyltransferases in the Golgi cisternae ensures their efficient access to sugar nucleotide donors and glycoprotein substrates and is critical for the proper glycosylation of glycoproteins and glycolipids. Our goal in this proposal is to elucidate the mechanisms responsible for the localization of Golgi glycosyltransferases. Recent observations suggest that anterograde cargo proteins (e.g. cell surface and secreted proteins) move through the Golgi, not in vesicles, but in cisternae that sequentially mature from the cis to the trans face of the Golgi stack. In this cisternal maturation model of protein transport, Golgi proteins are not retained in specific cisternae, but are transported via COPI vesicles in a retrograde fashion, thereby maturing the cisternae progressing through the stack, and consequently maintaining their steady state cisternal localization. In this proposal we describe experiments which will allow us to evaluate the role of COPI retrograde transport, protein oligomerization, and lipid partitioning in Golgi glycosyltransferase localization. We have chosen to study three late Golgi glycosyltransferases, beta 1, 4-galactosyltransferase-I (GaIT), alpha 2, 6-sialyltransferase (ST6Gal I), and alpha1, 2-fucosyltransferase (alpha1, 2-FucT), which have overlapping, but distinct, sequence requirements for localization. In the first aim we will ask (1) whether late Golgi glycosyltransferases are found in COPI transport vesicles, (2) what sequences mediate vesicle incorporation, and (3) whether COPI vesicle incorporation is necessary for enzyme Golgi localization. In the second aim, we will test the hypothesis that enzyme oligomerization plays a role in COPI vesicle incorporation and Golgi localization by determining (1) the enzyme sequences required for oligomerization, (2) whether enzymes in COPI vesicles are exclusively in an oligomeric state, and (3) whether cisternal factors that control enzyme oligomerization can be manipulated to reverse this process. The information obtained from this work will not only increase our understanding of the complex mechanisms underlying protein trafficking and localization in the Golgi, but will also allow us to evaluate how the organization of the glycosylation pathway impacts the types of glycans made by a cell.